ChatGPT Proves Effective in Cybersecurity Tests, Yet Raises Concerns Over Vulnerability Exploitation

A study by Charles Darwin University has found that Artificial Intelligence (AI) can help a lot in fighting cyber crime. The study was done by the professors from the Institute of Energy and Resources of CDU and they published their research in the International Journal of Information Security. The researchers from the university used generative AI for penetration testing known as pentesting. Pentesting is often used by cybersecurity to find weak spots in any system. Pentesting testing was performed on AI by giving it a few assessments and activities and then asking ChatGPT to identify threats. The testing included getting into a computing machine to download some files, analyzing their source codes in web pages, and finding previous or hidden flaws.

Dr. Bharanidharan Shanmugam, who took part in this research, said that they did this study to find if generative AI like ChatGPT can be useful to detect any harmful hacks and the results of the study were somewhat satisfactory. According to Shanmugam, ChatGPT is capable of detecting cyber crimes that are led by specific systems and organizations. During the scanning period, ChatGPT can easily do thorough scanning of the target and can then identify vulnerabilities and cyber crime activities in a system.

• Related: ChatGPT Use Ruins Memory and Academic Performance According to This Study By Pakistani Researchers

Even though ChatGPT has proved to be a great tool during the above phases, it is also capable of exploiting vulnerabilities of the machines. ChatGPT is a revolutionary tool for pentesting but it should also be monitored. If ChatGPT is to be used as a cybersecurity tool, it should be adopted after following some practices and guidelines. By using ChatGPT and other generative AI tools, we can maintain a safe digital environment.

Image: DIW-AIgen

Read next: 85% of People Make Health Decisions Based on YouTube Videos